JP2007018321A - Information processing apparatus and printer driver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To eliminate complicated processing for performing offset adjustment for location information between at an edge in the scanning direction of a recording area in one scanning and a mask pattern, in a recording apparatus which receives mask data and record data compressed with the mask pattern. <P>SOLUTION: When recording is performed using record data compressed with the mask pattern, the location of an edge in the scanning direction of a recording area in one scanning is adjusted based on the size of the mask pattern. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an information processing apparatus and a printer driver.

  A serial type ink jet recording apparatus ejects ink from nozzles of a recording head while moving a carriage mounted with a recording head in a direction (main scanning direction) different from the paper feed direction (nozzle arrangement direction). Form.

  This recording apparatus is provided with a print buffer (recording buffer) for storing recording data to be transferred to the recording head, and the memory capacity thereof can hold at least all the data to be recorded in one main scanning direction. A sufficient amount of memory is secured (print buffer for one band).

  For example, when a recording head having 128 nozzles arranged at 600 dpi in the radial scanning direction performs printing at 8 inch width at a resolution of 600 dpi in the main scanning direction, the print buffer required for one main scanning is 76800 bytes per color. Requires a print buffer.

  On the other hand, in order to realize high image quality, a conventional multi-droplet method in which a plurality of discharge amounts for each color is used for one discharge amount per color, or C (cyan), M (magenta), Y In addition to the four basic colors (yellow) and K (black), there are a wide variety of multicoloring techniques that add PC (photocyan) and PM (photomagenta) colors to enhance the gradation of photo images. It is popular. For this reason, the amount of memory required for the recording apparatus tends to increase.

  Further, in order to realize a high speed, it is also continuously attempted to reduce the number of scans necessary for printing one page by increasing the number of nozzles of each color. This also increases the print buffer.

  An increase in the print buffer size leads to an increase in printer production costs. However, in conjunction with the spread of the host device (host computer) in the market and the price reduction, the price reduction of the inkjet recording device market is also progressing rapidly. In particular, low-end ink jet printing apparatuses are remarkably low in price, and in order to realize this, an invention for reducing the memory size of an ink jet printer (making the print buffer size smaller than one main scan) has been made. ing.

  For example, the control of starting main scanning before completing the input of data for one main scanning to the buffer (Patent Document 1) or the multi-task processing of the CPU in the host device from the host device due to an increase in processing load, etc. There is recovery control (Patent Document 2) when data transfer to the recording apparatus is not in time.

Furthermore, there is control (Patent Document 3) in which mask compression is performed in the host device, and information used for mask compression and print data that has been thinned out are transferred from the host device to the recording device.
JP 11-259248 A JP 2003-305903 A JP 2003-159839 A

  However, in the case of Patent Document 3, mask pattern data and compressed thinned data are sent in one column of data, and the mask pattern is a one-dimensional mask along the nozzle row. Moreover, the mask pattern is shorter than the length of one nozzle row and the mask pattern is repeated a plurality of times for the number of data corresponding to one nozzle row, which is advantageous in terms of cost. However, on the other hand, there is a problem that the degree of freedom of the mask pattern is reduced and the image quality is sacrificed.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a high-quality and low-cost recording apparatus while suppressing the amount of data transferred from the host apparatus to the recording apparatus.

  In order to achieve the above object, an information processing apparatus according to the present invention has the following configuration. That is, an information processing apparatus comprising: a first output mode for outputting recording data generated based on image data; and a second output mode for outputting the generated recording data after thinning out using mask data, Based on the selection of the first output mode or the second output mode and the selection of the selection means, the column position where the thinning by the mask data is started is changed.

  The printer driver of the present invention has the following configuration. That is, a printer driver having a first output mode for outputting recording data generated based on image data, and a second output mode for outputting the generated recording data after thinning out using mask data. Based on the selection of the first output mode or the second output mode and the selection of the selection step, the column position where the thinning by the mask data is started is changed.

  As described above, according to the present invention, it is possible to save the trouble of outputting position information for performing mask processing from the host device when recording is performed by compressing using mask data. Also for the printing apparatus, it is possible to omit a control configuration that receives position information for performing mask processing and performs position adjustment for performing mask processing.

  The basic principle of the present invention and preferred embodiments of the present invention based on this principle will be described in detail below with reference to the accompanying drawings.

<Overall configuration of printing system>
FIG. 1 is an external view of a printing system in an embodiment described below. This system includes a host computer 2001 and an inkjet printer (inkjet recording apparatus) 2011 connected to the host computer 2001 via a serial interface 170.

  FIG. 2 is a functional block diagram of the printing system. In the host computer 2001, the printer driver 2002 obtains data arranged in an RGB format as color information and a raster format as a data array via an operating system graphic driver interface (GDI) 2004 from an image created by an application. To do.

  The printer driver 2002 executes processing (input correction, color correction, output correction, binarization, etc.) independent of the model characteristics of the inkjet printer 2011 that is the output destination, and execution of processes dependent on the model characteristics in raster units. The data to be transmitted to the inkjet printer 2011 is generated repeatedly. In particular, the model-dependent processing is executed by the output module 2003, which is one component module included in the driver 2002.

  In the output module 2003, in addition to the level conversion processing conventionally performed, the raster / nozzle conversion for converting from the YMCK raster data format to the format suitable for the later-described nozzle layout, which has been performed on the printer side until now. (HV conversion), print path control based on a combination of generation of recording data (print data) and paper feed control, and scan control for detecting the left and right print ends of each scan are also executed on the output module side.

  As a result, the control circuit of the ink jet printer and the software executed therein can be simplified, which contributes to the realization of a low cost ink jet printer.

  Print data generated by the host computer 2001 is transmitted to the inkjet printer 2011 via the spooler block 2005 and the serial interface 170 of the host computer.

  In the ink jet printer 2011, data reception to the reception buffer 2 is controlled by a transmission / reception task 2012, a command stored in the reception buffer is analyzed by a command analysis task 2015, and control according to the command type is executed. . In the case of mechanical operation-related commands such as a paper feed command, a paper discharge command, and a paper feed command, an engine interface task 2017 that is an interface between the mechanical control system and the upper layer, an engine sequencer task 2018 that controls mechanical sequence control, a sensor / motor The mechanical operation is realized by the engine driver task 2019 that controls the input / output of the mechanical unit.

  If the command analysis task 2015 determines that the command is a decode command to be described later, a DMA transfer from the reception buffer 2 to the decode buffer 2020 is executed. If the command analysis task 2015 determines that the print information command (print left / right edge information, block size, number of blocks, number of color planes, etc.), the information is transmitted to the print data control task 2016. When the command analysis task 2015 analyzes the print data command, the command analysis task activates the DMA transfer of the print data from the reception buffer 2 to the recording buffer 4. Simultaneously with this activation, the print data control task 2016 instructs the engine interface task 2017 to start scanning in synchronization with a DMA transfer end interrupt for transferring print data of one block (for 64 columns).

  In the engine interface task 2017, after setting various print-related registers in the ASIC (so-called gate array) 174, the engine sequencer task 2018 and the engine driver task 2019 are used to perform carriage operation (print head operation). Scan). When the carriage reaches a predetermined encoder position, the ASIC 174 performs DMA transfer from the recording buffer to the recording head and driving control of the recording head based on the set value of the print-related register, and ejects ink from the recording head. Then, an image for one scan is formed on the recording paper (recording medium).

  Other tasks include a user interface task 2013 that controls detection of an input system operated by a user, such as a cover switch, a power key, and a resume key for detecting cover open / close, and a system for adjusting operations between the tasks. There is a control task 2014.

<Configuration of recording apparatus>
FIG. 3 is an external perspective view showing the mechanical configuration of the inkjet printer 2011. The ink jet printer has a configuration capable of both color printing and monochrome monocolor printing. As shown in FIG. 3, a multi-nozzle recording head 102 and a cartridge guide 103 having 320 nozzles for black and 192 nozzles for each color are mounted on the carriage 101. The recording head 102 ejects black (K) ink or cyan (C), magenta (M), yellow (Y), and black (K) ink, respectively. During the printer operation, the recording head 102 is mounted with an ink cartridge 110 containing black ink and an ink cartridge 111 containing other three colors of ink.

  Then, cyan (C), magenta (M), yellow (Y), and black (K) inks are supplied from the respective ink cartridges to the recording head. Further, a drive signal for each nozzle in the recording head is supplied via a flexible cable (not shown) in which a large number of conducting wires are arranged.

  FIG. 7A is an external perspective view showing the recording head 102 for color printing from the recording medium 106 side. The nozzle rows (yellow (Y) 701, magenta (M) 702, cyan (C) 703, black (K) 704) corresponding to each color are arranged in a direction substantially orthogonal to the scanning direction (X direction). . Each color nozzle of yellow, magenta, and cyan has a structure in which 192 nozzles are arranged in parallel with small nozzles and large nozzles as shown in the cyan nozzle layout diagram of FIG. 7B. Therefore, the total color is 1152 nozzles (192 nozzles x 6 planes (yellow large / small / magenta large / small / cyan large / small)). It is handled as a plane.

  The carriage 101 is mounted on two guide rails 104 and 105, and the endless belt 109 connected to the carriage 101 is driven by a carrier motor (described later) to move the carriage 101 in the X direction (hereinafter, this X direction is referred to as “ Reciprocating scanning in the “main scanning direction”. The conveyance roller 108 is driven by a conveyance motor (described later), and conveys the recording medium 106 in the Y direction (hereinafter, this Y direction is referred to as “sub-scanning direction”).

  An encoder slit (not shown) is arranged in parallel with the guide rail, and a sensor (not shown) mounted on the carriage 101 captures the position in the scanning direction by reading the number of encoder slits. Based on this, the position of the carriage is controlled in units of one pixel.

  FIG. 4 is a block diagram showing a control circuit of the ink jet printer. In FIG. 4, reference numeral 170 denotes an interface for inputting a recording signal. For example, recording data is input from an external device such as the host computer 2001. 171 is an MPU, 172 is a ROM that stores a control program (including character fonts if necessary) executed by the MPU 171, and 173 is a temporary storage of various data (such as the recording data and data supplied to the head). This is an SRAM.

  Reference numeral 174 denotes an ASIC (so-called gate array) that performs recording control on the recording head 102, and also performs data transfer control among the interface 170, MPU 171, and SRAM 173. 179 is a carrier motor for moving the recording head 102 in the main scanning direction, and 178 is a conveying motor for conveying the recording medium in the sub-scanning direction. Reference numeral 175 denotes a head driver for driving the recording head, and reference numerals 176 and 177 denote motor drivers for driving the transport motor 178 and the carrier motor 179, respectively.

  The outline of the operation of the control circuit block diagram described above will be described. When recording data enters the interface 170, the recording data is converted into driving data between the ASIC 174 and the MPU 171. Then, the motor drivers 176 and 177 are driven, and the recording elements of the recording head 102 are driven according to the drive data and control signals sent to the head driver 175 to eject ink.

<Data flow in the recording device>
FIG. 5 is an explanatory diagram of a control block of the recording apparatus. Hereinafter, the data flow in the recording apparatus will be described with reference to FIGS.

  Reference numeral 1 denotes an interface control unit (controller) which receives data transferred from the host computer 2001 via the interface signal line S1, and among the received data, data and image data necessary for the operation of the recording apparatus. Is extracted and stored once. The data extracted by the interface controller 1 is stored in the reception buffer 2 via the signal line S2.

  The reception buffer 2 is provided in a memory means (storage device) such as SRAM or DRAM, and the data stored in the reception buffer has a structure as shown in FIGS. 8 (a), (b), and (c). Become.

  As shown in the data structure of the reception buffer in FIG. 8A, data of “command” (1001), “data length” (1002), and “setting data” (1003) are stored in this order from the left. Subsequently, data of “command” (1004), “data length” (1005), and “setting data” (1006) are stored.

  This indicates that data transferred in chronological order is stored at successive addresses in the reception buffer. The setting data 1006 shown here includes, for example, execution of paper feed, paper feed amount setting, and recording head to be used. It is information indicating the number and the like, and recording is possible with the recording apparatus only after all the information determined by the setting data is prepared. Thereafter, the image data (1009, 1012) to be recorded is stored in the reception buffer 2.

  The image data (1009, 1012) is data obtained by dividing the amount of data required when the recording head performs recording on the recording medium in one scan into smaller blocks as a data amount. The image data is divided in units of blocks, and stored as first block data (1009), second block data (1012),. These block data are also stored in block units in the recording buffer.

  FIG. 8B is a diagram showing in detail the data structure of image data divided into block units. As shown in FIG. 8B, a plurality of color data (1013 to 1014) are sequentially compressed as data. Stored. This color data is delimited by “color changing codes” (1016, 1017, 1018).

  For example, in the case of a printing mode (printing mode) that is applied to plain paper and is a printing mode (recording mode) that uses only large nozzles for four colors of cyan, yellow, magenta, and black, the compressed first The data from one color to the fourth color constitute one data block.

  One of the modes applied to special paper is a printing mode (recording mode) that forms an image with three colors of cyan, yellow, and magenta dyes without using Bk, which is a pigment that is difficult to fix on special paper. is there. In this case, the graininess is reduced by using a small nozzle in addition to the large nozzle. That is, for three colors of cyan, yellow, and magenta, two nozzle rows (large nozzle and small nozzle) in the scanning direction, each having 192 vertical nozzles for each color, and these two nozzle rows are cyan and yellow, respectively. , Provided corresponding to magenta.

  That is, the compressed color data of the first to sixth colors is configured as one block data. For example, the first and second colors are cyan data, the third and fourth colors are magenta data, and the fifth and sixth colors are yellow data.

  FIG. 8C shows the details of the data structure stored in the reception buffer for one-scan printing in the printing mode in which the output module applies mask compression described later. The difference from FIG. 8A which is the data structure in the case where mask compression is not applied is that the scan information command (1021 to 1023) indicating the print left end position and right end position of this scan and the print image data command of the first block are as follows. (1030 to 1032), two commands for transmitting information necessary for decoding the image data mask-compressed by the output module on the inkjet printer side, a decode information command (1024 to 1026) and a decode data command (1027 to 1029) is added.

  The decode information command is a command for designating which decode pattern is used to decode each color in the print image data command. The decode data command is the decode data itself, and in this embodiment, a maximum of 4 patterns are supported. The decoded data is data used at the time of executing the mask compression executed by the output module, and will be described in detail in the mask compression by the output module described later, but forward printing (from the carriage reference position to the non-reference position). The data used for mask compression differs between the time of scanning and the reverse printing (scanning from non-reference to reference). For this reason, the output module generates a decode information command and a decode data command for each scan.

  Here, although the control block of the recording apparatus is being described, the recording buffer will be described. FIG. 6 shows the data structure of a recording buffer that holds image data. In this embodiment, the number of columns in one block is 64 columns (600 DPI), and the recording buffer has a ring buffer structure.

  For example, in a mode that is applied to special paper and uses both Y, M, and C large nozzles and small nozzles, the first color data to the sixth color data are stored in each block data. The The length of each color data stored in each block corresponds to the number of nozzles of the recording head. As the size of the recording buffer, three blocks of the maximum color planes that can be taken are allocated, and the maximum number of color planes that can be taken is one of the modes used for the special paper printing, and is the first color data to the sixth color data. 6 of the case. Accordingly, 27648 bytes (192 nozzles × 64 columns × 6 color planes ÷ 8) are allocated to the DRAM 123 as the recording buffer size.

  For example, when recording a maximum length of about 8 inches in one scan in a printing mode using 6 color planes, it is possible to hold 3 blocks at the same time as described above. Buffer A is assigned buffer B to the second block, and buffer C is assigned to the third block. When the reading (that is, printing) of the first block from the buffer A is completed, the data of the fourth block is written into the buffer A. When the reading of the second block from the buffer B is completed, the data of the fifth block is written into the buffer B. The rotation (ring buffer) as described above is repeated, and the data of the 75th block (8 inches × 600 DPI ÷ 64 columns) is finally written into the buffer C. The processing in such a recording buffer employs a configuration in which reading is performed in parallel with writing by managing write pointers and read pointers, which will be described later, thereby providing a width of one scan (scan width) of 64 columns × 3 blocks. Even in a recording buffer that is much less than 1, it is possible to perform one-scan printing.

  In another case, for example, when only one color is used, it is possible to hold 18 blocks of data at the same time. However, even in this case, the print width is 1.92 inches, and the maximum print width is The recording buffer size is far less than an 8 inch.

  Here, the description returns to FIG. 5 to continue the description of each control block. Of the data stored in the reception buffer 2, “command”, “data length”, and “setting data” which are setting values for controlling the recording apparatus are read from the interface controller 1 by the CPU 9 via the signal line S 902. Are set in the respective control circuits (7, 8) in the figure (S903, S907). The CPU 9 interprets the read data (data corresponding to 1001 to 1008 in FIG. 8A), and performs overall recording control of the recording apparatus according to the result.

  On the other hand, regarding the processing of the image data, the CPU 9 activates the data expansion block 3 to execute the processing. The data development block 3 has three functions. The first function is a download function for downloading decoded data. The second function is a decompression function that simply decompresses (decompresses) PackBits of print data compressed in PackBits. The third function is a decompression function that sequentially performs a decompression process using decoded data and a PackBits decompression process for double-compressed data such as mask compression and PackBits compression. This decompression process is an expansion process if another expression is used.

  The PackBits expansion performed in the data expansion block when mask compression is not performed as shown in FIGS. 8A and 8B will be described. The data expansion block 3 reads out three types of data “compression TAG”, “data”, and “color change code” from the reception buffer 2 as shown in FIG. 8B, and expands the data based on these data. Execute control. If the compression TAG is 8 bits and the value is 00h to 7Fh, it is processed as if there are 1 to 128 non-contiguous data in the data area. If the compression TAG is 8 bits and the value is FFh to 81h, the next 1-byte data Is performed to decompress the data into continuous 2 to 128 data. When 80h is read at the compression TAG, it is processed as a color change code. The decompressed data is placed on the signal line S4 and written to the recording buffer 4.

  The decompressed image data is stored in the recording buffer 4 as shown in FIG. At the head address (head address) of the recording buffer 4, the head data of the data corresponding to the first color of the first block is written (stored). Subsequent data is sequentially written while adding addresses one by one. The area that can be stored in the recording buffer is determined by the setting data read by the CPU 9 first, and data larger than that value cannot be written. Therefore, when compressing image data, the data size is restricted according to the setting data. Will be. The data after detecting the color change code is sequentially written from the top address of the data corresponding to the second color. The control of the address data is executed by a recording buffering control structure circuit 8 to be described later.

  This writing process is performed from the first color data to the eighth color data in the first block, and when the color change code is detected after the eighth color data has been written, all the data in the first block has been written. . The data expansion block 3 finishes the data expansion operation, notifies the CPU 9 of the completion of data expansion for one block by an interrupt (INT1), and waits for the next data expansion from the CPU 9 to start.

  When a plurality of blocks of image data are prepared in the recording buffer 4, the CPU 9 operates the carrier motor 179 to start the recording operation, and the recording head 6 scans the image data in synchronization with the carriage encoder (CR encoder) 10. Then, the image can be completed on the sheet (on the recording medium) by transferring and recording. After the recording head 6 scans in the main scanning direction, the conveying means conveys the recording medium in the sub scanning direction. In this way, scanning of the recording head and conveyance of the recording medium are repeatedly performed to record an image for one page.

  The recording data generation block 5 reads the recording data in the recording buffer 4 through a signal line S5 at a timing synchronized with the CR encoder 10 in accordance with a value designated by the CPU 9, and converts it into a data structure that can be recorded by the recording head 6. The signal is output to the signal line S6. This recording data generation block 5 holds information on the block width (indicating the length of the block) in the recording buffer, which will be described later, and information on the height of each color of the block (referred to as “raster number” of color data). To do.

  Next, a decoding data download function applied to the decoding data command (1027 to 1029) shown in FIG. 8C will be described. The data expansion block 3 reads the decode data command (1027) from the reception buffer 2 and reads the data length (1028), and directly decodes the decoded data fetched via the signal line S3 to the signal line S2001 without performing expansion or decompression processing. The data is transferred and DMA-transferred to a decode buffer assigned to a part of the DRAM 173. In this embodiment, four patterns of decoded data from pattern 0 to pattern 3 are stored, and an area of 1536 bytes (vertical 192 rasters × 64 horizontal columns ÷ 8) is assigned to each pattern.

  Finally, as shown in FIG. 8C, the mask development executed in the data development block when the decode information command (1024) is detected between the scan information command (1021) and the print data command (1030). explain.

  Before starting the PackBits development, the CPU 9 determines the decode pattern No to be applied to the data corresponding to each color by the parameter (1026) of the decode information command (1024), and masks the data development block via the signal line S905. In a development register group (not shown), the mask development function is enabled, the decode pattern No. applied to each color data, and offset information are set.

  Each decoding pattern is two-dimensional data of 192 rasters × 64 columns, and the offset information specifies from which position decoding is to be started. Thereafter, similarly to the PackBits expansion, the CPU 9 activates the PackBits expansion by the data expansion block via the signal line S905. The data expansion block 3 takes in the image data of the reception buffer via the signal line S3 and executes PackBits expansion. At the same time, the decode data corresponding to this data is obtained from the decode buffer via the signal line S2002, the data before mask compression is restored from the data after the PackBits expansion and the decode data, and then the recording buffer is obtained via the signal line S4. 4 stores the recording data.

<Receiving buffer write / read control>
The interface controller 1 writes data to the reception buffer 2 and the data development block 3 reads only image data. The reception buffering structure control circuit 7 controls the write address and read address. The reception buffering structure control circuit 7 manages the start address and the end address of the reception buffer 2, and the write address and read address.

  The reception buffering structure control circuit 7 adds a write request signal (S701) received from the interface controller 1 for each reception, and outputs this to the reception buffer 2 as write address information (signal line S702). The reception buffering structure control circuit 7 performs control to return the write address to the head address of the reception buffer 2 when the final address of the reception buffer 2 is reached.

  When the write address reaches (matches) the read address, the reception buffer 2 is filled with data and communicates to the interface controller 1 via the signal line S703 that the next data cannot be written.

  At the same time, the reception buffer 2 notifies the CPU 9 that the data cannot be written by an interrupt signal on the signal line S904. The structure of the reception buffer 2 can be set by the CPU 9 writing to an internal register using the bus of the signal line S903.

  When the CPU 9 reads the data in the reception buffer 2 directly through the data read register in the reception buffering structure control circuit 7, the data expansion block 3 sends the data read request signal line S705 to the read address. When the request is made, the read address is added one by one via the signal line S706 and outputted to the reception buffer 2.

  The reception buffering structure control circuit 7 performs control to return the read address to the start address of the reception buffer 2 when the read address reaches the final address. Further, when the read address reaches (matches) the write address, data disappears from the reception buffer, so that the next data cannot be read is communicated to the data development block via the signal line S704. At the same time, the CPU 9 is informed that there is no data to be read out on the reception buffer 2 through the interrupt signal line of the signal line S904.

  The above is the processing content of the data writing / reading control to the reception buffer 2. Next, the processing contents for writing the data read out from the reception buffer 2 and developed into the recording buffer or reading the data from the recording buffer will be described.

<Record buffer read / write control>
The data development block 3 writes image data to the recording buffer 4, and the recording data generation block 5 reads the written image data. At this time, the recording buffer controls the writing address and the reading address. This is a ring structure control circuit 8.

  The recording buffering structure control circuit 8 manages the start address, the end address, the write address, and the read address of the record buffer.

  The recording buffering structure control circuit 8 adds a write request signal (S801) received from the data development block 3 by one address each time it is received, and outputs this to the recording buffer 4 as write address information (signal line S802). Then, the recording buffering structure control circuit 8 performs control to return the write address to the head address of the recording buffer 4 when the final address of the recording buffer 4 is reached.

  When the write address reaches (matches) the read address, the recording buffer 4 is filled with image data, and the fact that the next image data cannot be written is communicated to the data development block 3 via the signal line S803.

  When the data expansion block 3 reads the color change code from the reception buffer 2, the data expansion block 3 communicates that fact via the signal line S804, and the recording buffering structure control circuit 8 receives the next color data. The head address to be stored is prepared to be output from the signal line S802. The structure of the recording buffer 4 can be set by the CPU 9 writing to an internal register using the bus of the signal line S907.

  When the recording data generation block 5 requests the read address for each color via the data read request signal line S805, the read address is added one address at a time via the signal line S806 and output to the recording buffer 4.

  The recording buffering structure control circuit 8 performs control to return the reading address to the head address of the recording buffer 4 when the reading address reaches the final address.

  The recording data generation block 5 sets the data structure of the image data block currently being read out from the CPU 9 to a register in the recording data generation block 5 via the bus of the signal line S908. When all the image data in the set image data block structure is read, the end signal S909 is communicated to the CPU 9 as an interrupt signal. At this time, if the next image data block is already developed on the recording buffer 4, the image data block structure is written in the register.

  Since the recording buffer 4 controls data writing in units of one image data block and does not activate a recording data generation block for an unwritten image data block, the read address of the recording buffer exceeds the write address. Absent.

  The above is the outline of the data flow in the recording apparatus.

<Data generation in the host device>
In the printing system of this embodiment, data generation in an output module, which is one of the modules constituting the printer driver executed on the host device side, will be described. This will be described with reference to FIGS.

  FIG. 11 shows a control flow of the output module.

  An output module 2003 acquires one raster data of each color of C, M, and Y from other modules of the printer driver 2002. (S3001) In the case of this print mode, as shown in (a) and (c) of the level conversion & random index image diagram of FIG. 9, it is delivered as 9 levels of information with 4 bits per pixel at a resolution of 600 DPI. . When there is data of 8 inches which is the maximum print width, the information amount is 2400 bytes (8 inches × 600 DPI × 4 bits ÷ 8) per color.

  In response to the input from the driver, as shown in the input / output distribution table of FIG. 11 (c), in the output module, the combination of large nozzles and small nozzles of each color is 4 bits according to the input values from 0 to 8. Sort to output. Among these, for those for which three patterns are assumed for the same input value, random sprinkling with random numbers is performed.

In the case of this printing mode, color 6-pass printing and color 192 nozzles make the paper feed amount 32 rasters. Therefore, every time 32 raster data is accumulated in the random index buffer, data generation for one scan is executed. When the number of rasters accumulated after paper feeding is less than 32 rasters, the processes of S3001 and S3002 are repeated, and the sprinkling process using the random index is repeated. (S3003)
If 32 new rasters are accumulated in the random index buffer, the process proceeds to various data generation processes for this scan. As a more specific example, therefore, data generation for performing the P1 scan of the 6-pass image in FIG. First, data to be printed in the main scan is extracted from the random index buffer to generate an intermediate buffer. (S3004)
As shown in the relationship between the output bit position and the pass in FIG. 9D, each 4 bits on the random index buffer are defined so that 1XXX is printed at 1 and 4 Pass, X1XX is printed at 2 and 5 Pass, and XX1X is printed at 3 and 6 Pass. Has been. That is, in the P1 scan in the 6-pass image diagram of FIG. 10F, the data for printing with the nozzle block A (32 nozzles counted from the paper feed side) is N blocks in which the current 32 raster data is stored. Among them, 1XXX (first bit of 4 bits) is extracted. Similarly, in the P1 scan, the data to be printed in the nozzle block B (32 nozzles following A) is X1XX (second bit in 4 bits) in the (N-1) block in which the previous 32 raster data is stored. ) Will be extracted. For nozzle block C, N-2 block XX1X, for nozzle block D, N-3 block 1XXX, for nozzle block E, N-4 X1XX is on the most downstream side (discharge side). For the nozzle block F, which is 32 nozzles positioned, XX1X of N-5 blocks is extracted.

  That is, in FIG. 10 (f) 6-pass image diagram, intermediate buffer data generated for printing with the nozzle block A in the forward path of the P1 pass and intermediate buffer data generated for printing with the nozzle block D in the return pass of the P4 pass. Are basically the same data, but are strictly different. The reason is that, as described in the flow of recording data in the printer main body, the recording buffer on the printer main body side can hold only print data for 64 columns × 3 blocks simultaneously in the color 6 color plane mode. This means that the search direction of the random index buffer based on the extraction pattern depends on the scan direction (forward direction printing (Fwd) / returning direction printing (Bwd)) as shown in the random index buffer of FIG. This is because it must be switched. The above is described only for one color plane. However, since there are large nozzles and small nozzles for the three colors Y / M / C, the intermediate buffer is similarly changed from the random index buffer to the total six color planes. Will be generated.

  The intermediate buffer of all color planes is searched to detect the used nozzle width (S3005), and the leftmost print end and rightmost print end in all color planes are detected (S3006).

  In this figure, print data of 50% Dot (dot) is generated in the P1 forward pass, and print data of 50% Dot is generated in the P2 backward pass.

  Next, data generation of the mask buffer will be described. In the case of mask compression applied in the YMC large / small nozzle 6-pass mode of this embodiment, as shown in FIG. 10 (d1), the vertical is 192 rasters of the number of color nozzles, and the horizontal is 64 columns (both 600 DPI). (Unit) two-dimensional pattern mask buffer is applied.

  As for the contents of the mask pattern, as shown in the enlarged view of the mask buffer in FIG. 10 (d2), the mask buffer data is 50% Duty and the complementary relationship is established for the nozzle blocks having the same extracted pattern. And In the enlarged view of the mask buffer, ◯ means dots that do not transfer print data in the current scan, and ● means dots that transfer print data in the current scan.

  Mask A which is mask data for nozzle block A (32 nozzles counted from the paper feed side, that is, upstream in the transport direction), and nozzle block D (97th nozzle, 128th nozzle from the paper feed side, that is, upstream from the transport direction) (D2) As shown in the enlarged view of the mask buffer, the ● portion of the mask A is marked as ◯ in the mask D and the mask A as shown in FIG. The circled portion is marked with ● in the mask D. Similarly, the mask B and the mask E have a complementary relationship, and the mask C and the mask F have a complementary relationship.

  The above-described mask buffer of 192 raster lines and 64 columns is generated for all color planes (6 planes of Y, M, C, y, m, and c) used in the current print mode. (S3007) As described above, the search direction of the random index buffer for generating the intermediate buffer needs to be switched in accordance with the scan direction. This also applies to the mask buffer data generation. Depending on the scanning direction, it is switched. However, in FIG. 10 (c2) intermediate buffer enlarged view, (d2) mask buffer enlarged view, and (e) print data enlarged view after mask compression, in order to facilitate explanation of mask compression, the forward printing of P1 scan is performed. Data generation for printing with the nozzle block A and data generation for printing with the nozzle block D for P4 scan return pass printing are expressed as having no difference in the scan direction.

Next, mask compression processing is performed, but before that, rounding is performed on the mask block width at the leftmost end of printing and the rightmost end of printing detected in S3006, which is the point of the present invention. (S3008)
FIG. 12 shows the positional relationship between the image S recorded on the recording paper and the print leftmost end and the print rightmost end of the recording head for recording on the star-shaped image S. Further, on the upper side of the star-shaped image S, a print buffer for storing print data for one scan and a scan area of the print head are associated with each other, and the print data stored in the print buffer is described. Supplementally, in this embodiment, as described with reference to FIG. 6, the capacity of the recording buffer is smaller than the amount of recording data used in one scan, but the position in the scanning direction (column) recorded on the recording medium. In order to make it easy to understand the relationship between the (position), the state stored in the recording buffer (address of the recording buffer), and the mask data, they are represented as shown in FIG.

  When forming the image S on a recording medium, it is explanatory drawing about the scan of 6 times (P1-P6 is a scanning start position). For example, the first scan (P1) scans in the right direction for recording. In the second scanning (P2), the recording is performed by scanning in the left direction. Thus, bidirectional recording is shown. In FIG. 12, the recording paper is conveyed, but for the sake of simplicity, the recording head is shown shifted in the conveying direction.

  In this embodiment, 192 raster × 64 columns of two-dimensional mask data from address Pos0 in the buffer corresponding to the leftmost printable to address Pos75 corresponding to the rightmost printable (when the maximum print width is 8 inches). It is repeatedly applied in units of 64 columns.

  When performing mask compression, mask data corresponding to the print data position is picked up from the mask buffer and compressed. The mask-compressed print data is transferred from the host device to the printer before scanning as the mask data used for the mask compression, and the printer expands the mask from the mask-compressed print data and the decoded data (restoration from mask compression). )It is carried out. Therefore, if the relationship between the print data and the mask data at the time of mask compression on the output module side does not completely match the relationship between the print data and the decode data on the printer body, the original print data is completely restored. I can't.

  If the present embodiment is not applied, information about what column on the mask buffer (decode buffer on the printer body side) the print start position corresponds to, that is, offset information is required. As described above, it is necessary to switch the data generation of the mask buffer according to the scanning direction, and accordingly, the management of offset information becomes more complicated. For example, if each color plane (in this case, 6 color planes) has a decoding buffer that can hold 6 patterns that can be set completely independently, it is possible to eliminate the offset information. The number of ASIC gates on the printer main body side increases.

  As described above, by applying this embodiment, in the case of a mode in which mask compression is applied, the offset information is converted to the output module side by rounding the print start position to a value divisible by the column width of the mask buffer. Therefore, the control on the host device (printer driver or output module) side and the control on the printer side can be greatly simplified.

  Whether or not to execute this rounding process is whether or not the recording mode uses the mask data for compression. When the recording mode does not use mask data, it is not necessary to perform such rounding processing. This is shown in the control flow of FIG. In step S401, it is determined whether the recording mode is a mode using mask data. If Yes in step S401, processing for changing (rounding) the end position in the scanning direction in the print area (upstream end, downstream end, left end, and right end in the print head movement direction in the print area) is performed. . This is to change the reading start address of the recording buffer corresponding to the end position in the scanning direction in the print area, if the expression relating to the reading process of the recording buffer is used. The change of the read start address is to match the read start position with the address position delimited by the mask buffer. Then, when the read start position is aligned with the address position delimited by the mask buffer, for example, the recording data storage position is included in FIG.

  However, if No, the end position change (rounding) processing is not performed. The rounding method of the printing start position is switched according to the scanning direction as shown in FIG. For example, in the case of printing in the forward scan indicated by P1 in FIG. 12, the original print left end position (start position) is at a position (PosX + 3) advanced by two columns from PosX + 1, which is the 64-column separation position of the mask buffer. This position is changed from PosX + 3 to PosX + 1 in the print leftmost position in the forward scan of P1 (to the left in the figure). In the case of printing in the backward scan indicated by P2, the printing right end position (start position) shown in the figure is changed to PosY + 4 (to the right in the figure).

  This change makes it possible to extract (read out) mask data from the first column of the mask buffer without any management of the offset amount.

  Although the scan width (scan width) of the recording head for two columns is increased, this time is sufficiently negligible in the multi-pass print mode in which an image is formed in six passes.

  For example, when printing speed is required in a mode that does not apply mask compression, such as the printing mode for plain paper, rounding in units of column width (64 columns) of the mask buffer in S3008 is not executed. .

  When the generation of the mask buffer data is completed, the process proceeds to a process for generating print data compressed with mask from the print data extracted on the intermediate buffer and the mask data generated on the mask buffer. First, a raster pointer that is a pointer in the sub-scanning direction indicating which nozzle block (32 nozzle width) of 192 nozzles is being processed, and a column pointer that indicates a data pickup position in the main scanning direction from the intermediate buffer and mask buffer Initialize various pointers. (S3009) The raster pointer is a pointer for managing the data pickup position in the sub-scanning direction from the intermediate buffer and the mask buffer. When the raster pointer is 0, the nozzle block A (32 nozzles counted from the paper feed side) Raster position for the nozzle block B (32 nozzles following nozzle block A) when the raster pointer is 32, and the nozzle block F (if the final value is 160). The raster position for 32 nozzles following nozzle block E) is shown. The column pointer is initialized with values corresponding to the print leftmost and rightmost positions determined in S3006 and S3008. When the current scan is forward printing, it is determined from the leftmost position of printing, and when the current scan is backward printing, it is determined from the rightmost position of printing.

  In the print mode in which rounding is performed with the column width of the mask buffer in S3008, Null data (from the rounded leftmost print position or rightmost print position to the actual print leftmost position or rightmost print position is Null ( Null data, that is, zero data) is generated. That is, when generating block data including a print start end, null data for a column to be changed is inserted before data corresponding to the print start end.

  The column pointer also indicates the pick-up position in the main scanning direction of the mask buffer. As described above, since the mask is repeated at a period of 64 columns, the remainder is obtained by dividing the column pointer by 64 which is the column width of the mask buffer. The column position indicates the data pickup position with respect to the main scanning direction from the mask buffer.

  As shown in FIG. 10C, the intermediate buffer enlarged view, the data of 32 rasters × 16 columns specified by the raster pointer and the column pointer are cut out from the intermediate buffer (S3010), and similarly, FIG. As shown in FIG. 4, data of 32 rasters × 16 columns corresponding to the cutout position on the intermediate buffer cut out in S3010 is cut out from the mask buffer (S3011).

  As shown in the enlarged view of print data after mask compression in Fig. 10 (e), only the part marked with ● is extracted from the print data in the intermediate buffer to generate masked data, and printed from the host PC to the printer The data is temporarily stored in an output buffer (not shown) for transmitting data. (S3012) That is, the intermediate buffer data corresponding to the position of the circle in the mask buffer is thinned out in this scan.

  By repeating the processing of S3010 to S3012 for six blocks from A to F in units of 32 nozzles (S3013, S3014), print data that is thinned by 50% for 192 nozzles for the current 16 columns is generated. Will do.

  When the data thinning process by mask compression for 192 nozzles × 16 columns is completed, the raster pointer is initialized and the column pointer is updated by 16 columns to execute the next 16 columns (S3016).

  When the data thinning process by mask compression is completed for all the columns up to the leftmost end of printing and the rightmost end of printing, the process from S3009 is repeated for the remaining five color planes. Output buffer data is created for the six color planes (S3017).

  PackBits compression is executed on the output buffer data in which the data after data thinning by the mask compression is stored. (S3018) By the control flow from S3010 to S3016, 256-bit data, which is 50% thinned out from 512 bits of 16 columns × 32 rasters, is stored in the Nth color output buffer. Stored repeatedly in the order of block A, B, C, D, E, and F.

  On the other hand, as shown in the recording buffer data structure in FIG. 6, the printer main body is provided with a recording buffer for data of the first color to the sixth color for 3 blocks with a width of 64 columns. Therefore, 192 nozzles × 64 column width data is picked up from the output buffer for the first color, and PackBits compression is performed. Next, data corresponding to the width of 192 nozzles × 64 columns is picked up from the output buffer for the second color, and PackBits compression is performed. This is repeated up to the output buffer for the sixth color, and as shown in (b) of the received data structure in FIG. 8, a color change code is added to the end of each color data compressed in PackBits.

  The generation of the image data of the first block indicated by 1032 in (a) of the reception data structure in FIG. By repeating this process in units of blocks from the leftmost end of printing to the rightmost end of printing, image data of all blocks to be printed in the current scan is generated (S3020).

  Next, decode data to be transmitted at 1029 in (c) of the reception data structure of FIG. 8 is generated from the mask data generated at S3007. (S3021) In this embodiment, in the 6-pass mode in which thinning by mask compression is applied, the first color is cyan (large), the second color is cyan (small), the third color is magenta (large), The fourth color is defined as magenta (small), the fifth color is defined as yellow (large), and the sixth color is defined as yellow (small). The mask data for cyan is common to both large and small, and the pattern 0 is common to the mask data for magenta. The pattern 1 and the yellow mask data are the same in size, and the pattern 2 is applied (not shown).

  As described above, data thinning by mask compression was performed by repeating the process from nozzle A to nozzle F in units of 16 columns × 32 nozzles. Therefore, the decoded data generated from the mask buffer is also repeatedly repeated from nozzle A to nozzle F in units of 16 columns × 32 nozzles from the top of the mask buffer, and data for a total of 64 columns is obtained. Prepare to generate as one decoded data. As described above, the mask data stored in the mask buffer does not change depending on the scan start position, but changes depending on the scan direction (forward printing or reverse printing).

  This completes the data generation process for the current scan, and then shifts to a command / data transmission process from the host PC to the printer.

  First, the print leftmost position of the main scan determined in S3006 and S3008. A scan information command including information on the rightmost position of printing is transmitted. (S3022) Next, a decode information command is generated to notify the printer side which mask data each color plane is compressed (S3023).

  In the case of the 6-pass mode to which the mask compression described here is applied, as described above, the first color decoding pattern of 1034 in FIG. 8C is the pattern 0, and the second color decoding pattern of 1035 is also the pattern. 0 will be transmitted for pattern 3 for the third and fourth colors, and pattern 2 for the fifth and sixth colors.

  Next, the decoded data generated in S3021 is transmitted. (S3024) 192 raster × 64 column data, which is the mask data used for the mask compression for cyan, is transferred to the pattern 0 of 1037 in FIG. 8C, and the mask compression for magenta is transferred to the pattern 1 of 1038. The mask data used is the mask data used for the mask compression for yellow, to the pattern 2 of 1039, respectively.

  Finally, the image data generated in S3020 is repeatedly transmitted as a print data command in units of blocks, and the current command / data transmission process for scanning is completed (S3025, S3026).

  When the scan command / data transmission is completed, a paper feed command for 32 rasters is issued (S3027), and 32 rasters in the last random index buffer are released. For example, in the random index buffer of FIG. 10A, if the 32 rasters generated this time are the Nth, the last 32 rasters released this time are the N-5th 32 rasters. Actually, the random index buffer has a ring buffer structure, and the released 32 rasters branch to A in FIG. 11 and are used for processing for the next 32 rasters.

<Other embodiments>
In the above-described embodiment, the recording operation of the recording apparatus has been described by taking the case of bidirectional recording (forward recording and backward recording) as an example, but the present invention may also be applied to the case of unidirectional recording. As will be described with reference to FIG. 12, when the recording head performs only the forward scanning, the rounding process may be performed on the left end.

  Further, the number of columns of the mask block described above is not limited to 64, and may be 128 or 32, for example.

Overview of printing system Printing system functional block diagram External perspective view of recording apparatus Control block diagram of recording device Functional block diagram of the recording device Recording buffer data structure External perspective view of recording head Receive data structure Level conversion & random index image Image of mask compression & path control Output module control flow Image of the positional relationship between the print left and right edges and the mask Control flow

Claims (4)

An information processing apparatus comprising: a first output mode for outputting recording data generated based on image data; and a second output mode for outputting the generated recording data after being thinned out by mask data,
Means for selecting the first output mode or the second output mode;
An information processing apparatus characterized by changing a column position where thinning by the mask data is started based on selection by the selection means.   The information processing apparatus according to claim 1, wherein the mask data is two-dimensional data of m lines × n columns.   2. The information processing apparatus according to claim 1, wherein when the second output mode is selected, null data corresponding to a change amount of a column position is added when generating recording data. A printer driver comprising: a first output mode for outputting recording data generated based on image data; and a second output mode for outputting the generated recording data after thinning out with mask data,
Selecting the first output mode or the second output mode;
A printer driver, wherein a column position at which thinning by the mask data is started is changed based on the selection in the selection step.

Images